The Potential Impact of Federal Funding Cuts on Access to Pre-Exposure Prophylaxis in Atlanta, Georgia: Geographic Modeling Study

Introduction

We have arrived at a point in the HIV epidemic in the United States where we arguably have all of the biomedical tools we need to end the epidemic [1]. There are multiple options for HIV screening, including self-testing [2]; multiple formulations and regimens for pre-exposure prophylaxis (PrEP) to prevent HIV infection; and effective treatments that result in full healthy lives for people living with HIV and, for people living with HIV who are virally suppressed, eliminating the risk of onward transmission through sex [3]. Despite these biomedical advances, annual HIV transmissions in the United States remain above 30,000 [4].

The realties of bringing these foundational tools to scale in a coordinated response has proven to be a substantial challenge of implementation and has resulted in increased attention to and funding for implementation science approaches [5-7]. Geographic access to HIV prevention services, including PrEP, is foundational to the “Prevent” pillar of the National HIV/AIDS Strategy [8]. We have previously explored geographic access to HIV prevention and care issues during a time of more certainty and consistency about the funding mechanisms for HIV prevention and care [9-11]. We have found that inequities in geographic access to HIV prevention and care services, and often, commute times disproportionately impact communities of color for whom car ownership may be less [12,13] and for whom public transportation access and route frequency may be more limited [14,15].

As we consider the structural and socioeconomic factors that impact geographic access to HIV prevention services, we are now in an era where federal funding for HIV services is declining or under threats of future reduction or elimination [16-19]. HIV prevention services in the United States are supported through a patchwork of federal funding mechanisms, including direct service delivery grants from the Centers for Disease Control and Prevention, funds administered through the Health Resources and Services Administration like the Ryan White HIV/AIDS Program, the Ending the HIV Epidemic initiative, Title X, Medicaid reimbursement for clinical PrEP services, and time-limited federal demonstration and implementation grants. These funding streams vary in stability, allowable services, and eligibility criteria, creating differential vulnerability to funding reductions across delivery settings. Although most PrEP prescriptions in the United States are provided through traditional health care clinics, community-based organizations (CBOs) contribute disproportionately to PrEP access among marginalized populations through direct provision, partnerships with prescribing clinicians, and intensive navigation services [20]. Given that many CBOs that provide HIV prevention services are supported, at least in part, by federal funding, it is important to consider how reductions in federal funding that lead to the closure of existing prevention providers might impact geographic access to HIV prevention services. Further, given the disparities in new HIV diagnoses related to acquisitions risks (eg, men who have sex with men), racial/ethnic minority groups (eg, Black and Hispanic or Latinx people) and geographic region (eg, the US South), it is important to assess whether gaps in geographic access to prevention services are differential by these population characteristics. To address these gaps, we used public data about the locations of HIV PrEP providers and previously described methods for estimating commute times to HIV prevention care services [10] to model the potential impact of federal funding cuts on geographic access to PrEP in Atlanta, Georgia.

Methods

We examined how the closing of CBOs that provide HIV prevention services may impact access to PrEP by both car and public transit across the four counties that encompass metro-Atlanta: DeKalb, Cobb, Fulton, and Gwinnett (all of which are “Ending the HIV Epidemic”–prioritized counties for HIV prevention).

Ethical Considerations

This study used publicly available, deidentified census and clinic location data. No human participants were involved, and therefore, no reviews or approvals by an ethics committee or institutional review board were required.

Clinic Data and Closure Scenarios

We identified all PrEP-providing clinics in the metro-Atlanta area as of August 2025 by using the Centers for Disease Control and Prevention’s National Prevention Information Network PrEP Locator directory [21]. Clinics in the database are screened for eligibility if they have at least one health care professional (eg, physician, nurse practitioner, physician assistant) who is qualified to prescribe PrEP and if they have confirmed that they actively prescribe it. In addition to geographic data, each clinic is categorized by its funding type (federally qualified health center, public health department, hospital, CBO, etc). Of the 71 clinics in metro-Atlanta, 12 were designated as a CBO. To simulate potential impacts of federal funding reductions that disproportionately affect CBOs, we created 3 separate closure scenarios. In each scenario, 3 CBOs (approximately 25% of all Atlanta-area CBO PrEP clinics) were randomly selected for closure, and analyses were rerun to estimate the resulting changes in geographic access.

Origin and Destination Specification

Census block groups (CBGs) served as the unit of analysis for travel time estimation. Origin coordinates were defined as the population-weighted centroid of each CBG, representing where residents are most likely to live. CBG-level sociodemographic data were obtained from the American Community Survey 5-year estimates [22]. CBGs were categorized by the racial or ethnic plurality of residents. Destination coordinates were the locations of PrEP-providing clinics identified above. This approach captures area-level variation in accessibility rather than individual travel patterns.

Travel Time Estimation

We estimated one-way travel times between each CBG centroid and the 10 nearest PrEP clinics by using the Google Maps Distance Matrix application programming interface accessed through R [9]. The application programming interface was used to calculate the shortest available route under two transportation modes: public transit and private vehicle. For both modes, travel times were calculated across 3 weekdays (Tuesday, Friday, and Saturday) and 3 time points (8 AM, noon, and 3:30 PM).

The shortest one-way travel time for each day-time combination was selected and then averaged to provide one estimate per mode for each CBG for each closure scenario. CBGs with no available public transit route to any PrEP clinic, defined as a walking distance of more than 30 minutes to the nearest transit stop, were coded as having no transit access. Transit time estimates included walking time to any transit stops, waiting time, boarding/transfer time, and time in transit.

Statistical Analysis

We used descriptive statistics to describe mean one-way travel times to the nearest PrEP clinic before and after simulated closures, separately for car and public transit. The mean change in travel time (minutes) was averaged across the 3 closure scenarios and compared using a 2-sided paired t test. CBGs were dichotomized into those that saw increased travel times compared to no change for both car and public transit. We then used simple logistic regression to model associations between race/ethnicity and increased travel times. All analyses and visualizations were conducted in R software (version 4.3.2; R Foundation for Statistical Computing).

Results

Descriptive Results

A total of 2466 CBGs across metro-Atlanta were included in the analysis. More than half of the CBGs (n=1361, 55.2%) had a White plurality. An additional 943 (38.2%) CBGs had a Black plurality, 155 (6.3%) had a Hispanic/Latinx plurality, and only 2 (<1%) CBGs had a plurality of another race or ethnicity. Under baseline conditions, all CBGs had access to at least one PrEP-providing clinic by car within 30 minutes, while only 1027 (41.6%) CBGs had access via public transit within 30 minutes. A total of 567 (22.9%) CBGs did not have access to PrEP via public transit.

Public Transit Changes in Access

Across the 3 simulated CBO closure scenarios (N=2466), 732 (29.6%) CBGs experienced longer average public transit times to the nearest PrEP clinic compared with current access. These CBGs represented approximately 1,024,900 residents or about 27.8% of Atlanta’s total population. The average change in one-way transit time was 1.2 (range 0.0-11.6) minutes, which was significantly longer than the baseline (P<.001). Seven CBGs lost access to a PrEP clinic or CBO via public transit in the modeled closure scenarios. Compared to CBGs with a White plurality, the odds of experiencing increased transit times under the CBO closure scenarios were significantly lower for CBGs with a Hispanic plurality (odds ratio [OR] 0.61, 95% CI 0.40-0.95) but no different for CBGs with a Black plurality (OR 0.99, 95% CI 0.81-1.20).

Drive Time Changes in Access

Across the 3 simulated CBO closure scenarios (N=2466), 1184 (48%) CBGs experienced longer average drive times to the nearest PrEP clinic compared with current access. These CBGs represented approximately 1,698,000 residents or about 46.4% of Atlanta’s total population. The average change in one-way drive time was 0.5 (range 0.0-6.4) minutes, which was significantly longer than baseline (P=.03). Compared to CBGs with a White plurality, the odds of experiencing increased drive times under the CBO closure scenarios were significantly higher for CBGs with a Black plurality (OR 1.37, 95% CI 1.15-1.63) but no different for CBGs with a Hispanic plurality (OR 0.90, 95% CI 0.64-1.28).

Discussion

Our research explores the potential impact of federal funding cuts on geographic access to PrEP in Atlanta, Georgia. Whereas previous research in this area was proposed to identify geographic areas within cities that would be important locational targets for new services to decrease travel times to care [23-25], we argue that this expansion-oriented framing must be complemented by models of the closing of existing facilities due to cuts in HIV prevention funding to better understand PrEP service availability and the risk of new HIV infections. Our data suggest that increases in travel times related to the closure of community providers were greater for people who used public transit, underscoring the importance of evaluating modal dimensions of transportation time to health care facilities. This is especially relevant given that communities most vulnerable to HIV are also most likely to rely on public transportation [12,26].

Although the differences in travel times might seem modest in some settings, such differences represent meaningful barriers to seeking care among marginalized groups already facing transportation disadvantages. A given level of change in commute time may result in different willingness to travel for people who use different modes of transportation or for people who do not have employment benefits that include paid time for seeking health care [26,27]. It is also important to acknowledge that HIV prevention service providers have both located their facilities in areas with high needs for prevention services, and clinics that provide services in areas with fewer community and economic resources are often set up to meet multiple care and social service needs [28]. Colocation of PrEP services with other health and social services may facilitate engagement in prevention by reducing logistical barriers, normalizing HIV prevention within broader care settings, and leveraging established trust between providers and communities. Such organizations spend decades building trusting relationships with communities, fostering trusting relationships [29,30]. Depending on which facilities close, it is foreseeable that populations with multiple medical and social service needs might lose access to a variety of services that would increase their vulnerability to HIV and other health threats.

Other aspects of our analysis reinforce the stakes in terms of health and equity considerations that would attend closures of existing PrEP-prescribing organizations. Although any licensed medical provider can provide PrEP, CBOs are disproportionately located in high-need areas and serve populations often excluded from traditional health care systems [31]. Thus, the loss of such facilities would create a double impact on PrEP accessibility: it might create a longer commute time to the nearest PrEP provider, and it might mean that the nearest PrEP provider would be less likely to have culturally competent services [32-34]. Further, it is unclear whether existing clinics that survive the closure would have sufficient clinical capacity to provide PrEP to more patients even if prospective PrEP users are able to commute longer distances to continue to access PrEP [35,36]. Lastly, while our analysis focused on neighborhood level demographics, we know that PrEP-seeking patterns differ across multiple identity groups. For example, White men who have sex with men are more likely to access PrEP though traditional health care systems [37], while Black cisgender women and other groups with high HIV transmission tend to rely more often on community-based and safety-net providers [38]. Therefore, the loss of CBO-based PrEP services may disproportionately impact more marginalized groups.

The eventual impact of closures on the coverage of PrEP and the impacts of limitations on PrEP services are not easy to predict with confidence. However, it is clear that there are limitations to how far people will drive to obtain services for nonemergent health problems (eg, preventive services); several publications suggest that 30 minutes is a common threshold of willingness [39,40]. In any case, distance to receipt of care has an important and converse relationships with receipt of nonurgent health care, including preventive care [41]. A recent analysis suggests that even small decreases in PrEP coverage among people with indications could have substantial impacts on new HIV infections and health care costs. For example, just a 3% reduction in PrEP coverage is forecast to result in over 8000 preventable infections, with lifetime medical costs for HIV care exceeding US $3.6 billion for those infections [42].

Our findings, while empirically derived and reflective of real-world challenges to transportation to PrEP care services, have several limitations. First, we present one-way estimates of transit times, so the actual additional transportation burden to seeking PrEP care would be double the one-way estimates presented in the analysis. Simulated closures may not reflect actual likelihoods of closure; in the absence of empiric, publicly accessible data about organization-specific funding portfolios, we were unable to weight closures by likelihood and instead selected organizations randomly in the simulation. Future work incorporating such data could identify especially high-risk providers and communities. Conversely, we did not consider the likelihood of other types of organizations that provide PrEP closing; so, our estimates of impact on PrEP service might not reflect the true additional travel burden associated with the closure of facilities. Further, our analysis did not collect new data on, or consider the service capacity of, clinics; so, even if potential PrEP users were willing to travel longer distances, it is possible that existing clinics might not be able to handle an increase in the patient load.

According to our data, the closure of even a small number of current PrEP providers would have important impacts on physical access to HIV prevention services and would be predicted to increase commute times, decrease engagement in PrEP care, and result in avoidable new HIV infections and associated care costs. Therefore, our findings support sustained investment in CBOs as critical access points for HIV prevention services.